In an article originally published on the PSA’s political insight blog, Crick Centre Deputy Director Matt Wood examines the perils of punditry for political scholars
A debate has broken out in political science about how much we should be attempting to ‘self-promote’ via social media and other public forums. On the one side are the guardians of intellectual purity railing against all these ‘hot takes’ and ill-judged punditry that demeans our profession. On the other side are the hyper-active media superstars who defend the need to be ‘visible’ and ‘relevant’ to current events, and tend to monopolise media and public attention.
Given I myself wrote a blog before the general election assuming Labour would lose, I have been reflecting on my own shortcomings, and the wider role of our discipline in public life. I believe we need to reflect not only on whether we should be ‘self-promoting’ on the media or not, but also on the content of what we say. This is linked to a wider debate about whether political science should be about explanation or prediction, and I think if we’re careful we can make room for both, whilst avoiding the perils of punditry we all disagree with.
Guardians versus Superstars
One way to parody this debate would be to say it’s all fluff. This approach would say we should ‘just stop it’ and be getting on with the hard graft of theoretical introspection and data analysis, eschewing such trivial ‘punditry’. We simply just ‘do’ our analysis and then go about disseminating our research ‘findings’ and arguments through the usual channels. The media ‘superstars’, on this view, are simply careerists who attract attention by being ‘on call’ and accessible at any moment. Because of the vagaries of everyday news events, they end up commentating on things they often know little or nothing about. This in turn demeans our profession and makes us no different from political commentators aiming to push an agenda (often themselves).
While sympathetic to this perspective, I don’t think this debate can be dismissed so easily, because it’s linked to a much more substantive issue about what we can and should be claiming are legitimate insights of our own political scientific research. Given the ‘impact agenda’ is here to stay, it seems more productive to look at it from this angle than simply dismiss the superstar pundits.
Expanation and Prediction
A recent symposium in Political Studies Review journal debated a major new political science textbook by Prof Keith Dowding, which proclaimed prediction as the holy grail – and indeed the only legitimate final aim – of political science. Political scientists should, he argues, always be concerned with achieving prediction in their work. Tentative prediction, yes, but prediction that can be verified through systematic hypothesis generation and testing. Anything else, he argues, is either a mixture of political science and humanities-style research, or not properly ‘professional’ political science at all.
Contributors to the debate, in particular Prof Colin Hay, took umbrage at this claim. Hay argued that explanation – not prediction – ought to be the aim of an inclusive and legitimate political science. Social sciences claiming predictive capacity, on Hay’s view, can fall into quite sticky situations, either by failing to see exceptional events coming, or through the unexpected consequences when they gain unwarranted status among policy-makers or the public. Mainstream economists for example had unwittingly led policy-makers to become entranced by their ‘predictive’ capacity, leading eventually to catastrophic policy failures creating the global financial crisis. While Dowding continues to insist in his response that prediction is important, he concedes that some political scientists may want a different, more ‘retrospective’ approach to analysis.
Zoom over to the recent debate about political science and punditry after the UK general election and we can see this debate playing out. The ‘superstars’ predicted a heavy Labour defeat, and supported media punditry speculating on what the fallout would be. Re-tweeting polls as if they were hard and accurate predictions – even with protestations to the contrary about helping the public ‘understand’ the polls – created an atmosphere in which political science was seen as predictive, and evaluated as such. Unsurprising then, that once the mostly unexpected hung parliament happened, the intellectual guardians were quick to jump in to castigate those who had made it seem (again!) like the polls were accurate. The superstars created a febrile atmosphere in which political science actively influenced public expectations of the result, and, as in the Brexit case, led to turmoil in the markets and political tumult.
The superstars of course have come out defending the role of polls and data, and the necessity of public engagement, while the finger-wagging guardians say we should never be so bold as to claim predictive power for our findings. Both sides, in my view, could be reconciled if we readjusted our perceptions and practices of what public and media engagement is for, and what it is not for. If our presence in the media is a necessity, both from a moral perspective (we have a duty to the public to help them understand politics) and a practical one (it’s great for our careers and for more broadly shaping public and policy support for the discipline), then we need a debate less about whether we ‘should’ be visible, but the content of what we’re saying when we’re doing it. This has implications for what the public and media expect from us in practice, and could give us a better handle on how we influence public debate.
Explanatory and Predictive Public Engagement
I would suggest we divide our media and public engagement between two different roles: explanatory and predictive engagement. This would simply involve two equally legitimate forms of media/public engagement: predicting possible future political outcomes, and explaining previous political outcomes. Both can and should be placed within the context of existing political events. So for example, in the wake of the awful Grenfell tower disaster, political scientists should be concerned with:
1. predicting the implications of Grenfell for the distribution and exercise of political power, and:
2. explaining the political processes, events and long-run dynamics that led to Grenfell happening.
Both roles, in my view, are equally legitimate. Both also place important professional obligations on us to think about how we re-interpret our own work (which will necessarily be of limited direct application to the Grenfell disaster in itself), so as to contribute to media and public debate. However, it also places important and necessary limitations on us to not be led into stylized and often damaging ‘punditry’ that simply commentates on events as they occur. Our profession is not concerned with commentary simply for the point of it, but rather with explanation and prediction. The more we unnecessarily simplify the purpose of what we’re doing, on both counts, the more we devalue the distinctiveness of our profession.
All of this of course works in a broad political context. Real world events happen creating imperatives for engagement that cannot sensibly fit with long-term research outlooks. The most appropriate way to navigate this tension is not to compete in the journalistic punditry stakes, but to double down on our commitment to providing good background explanatory and predictive political analysis. The perils of punditry, in my view, can only be avoided by restating the basic facets of good scholarship, and looking to translate those into practice, rather than forgetting them altogether in the (unavoidable) race to be relevant.
Matt Wood is a postdoctoral research associate and Deputy Director of the Sir Bernard Crick Centre for the Public Understanding of Politics (Crick Centre). He has previously worked in local journalism and lobbying, and has held visiting fellowship positions at the UK Cabinet Office and ANZSOG Institute for Governance, University of Canberra.
Matt is responsible for the Crick Centre’s research strategy, and heads up the research strand on ‘Media, Science and Technological Change’. His research at the Crick Centre currently includes the study of everyday politics, celebrity politicians, the politics of normality, co-production as an approach to social research, and the concept ‘hyper-democracy’.
Matt is currently undertaking a three-year ESRC Future Research Leaders Fellowship, studying how delegated bodies can improve trust and support for their work from stakeholders in volatile political environments. This project will involve case studies in food safety, health technology regulation and disease control and prevention, and use an innovative ‘impact-led’ methodology including co-production.
This article was originally published on the PSA blog and is reprinted with permission of the author.
This article gives the views of the author, and not the position of the Crick Centre, or the Understanding Politics blog series.